All of lore.kernel.org
 help / color / mirror / Atom feed
From: Alberto Garcia <berto@igalia.com>
To: Stefan Hajnoczi <stefanha@redhat.com>
Cc: qemu-devel@nongnu.org, qemu-block@nongnu.org,
	Max Reitz <mreitz@redhat.com>, Kevin Wolf <kwolf@redhat.com>
Subject: Re: [Qemu-devel] [RFC] Proposed qcow2 extension: subcluster allocation
Date: Fri, 07 Apr 2017 14:24:35 +0200	[thread overview]
Message-ID: <w514ly0s8vw.fsf@maestria.local.igalia.com> (raw)
In-Reply-To: <20170407122021.GP13602@stefanha-x1.localdomain>

On Fri 07 Apr 2017 02:20:21 PM CEST, Stefan Hajnoczi wrote:
>> Here are the results when writing to an empty 40GB qcow2 image with no
>> backing file. The numbers are of course different but as you can see
>> the patterns are similar:
>> 
>> |-----------------+----------------+-----------------+-------------------|
>> |  cluster size   | subclusters=on | subclusters=off | Max L2 cache size |
>> |-----------------+----------------+-----------------+-------------------|
>> |   2 MB (256 KB) |  1200 IOPS     |  255 IOPS       | 160 KB            |
>> | 512 KB  (64 KB) |  3000 IOPS     |  700 IOPS       | 640 KB            |
>> |  64 KB   (8 KB) |  7200 IOPS     | 3300 IOPS       |   5 MB            |
>> |  32 KB   (4 KB) | 12300 IOPS     | 4200 IOPS       |  10 MB            |
>> |   4 KB  (512 B) |   100 IOPS     |  100 IOPS       |  80 MB            |
>> |-----------------+----------------+-----------------+-------------------|
>
> I don't understand why subclusters=on performs so much better when
> there's no backing file.  Is qcow2 zeroing out the 64 KB cluster with
> subclusters=off?

It is, see https://lists.nongnu.org/archive/html/qemu-devel/2015-07/msg05877.html

Berto

  reply	other threads:[~2017-04-07 12:24 UTC|newest]

Thread overview: 64+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-04-06 15:01 [Qemu-devel] [RFC] Proposed qcow2 extension: subcluster allocation Alberto Garcia
2017-04-06 16:40 ` Eric Blake
2017-04-07  8:49   ` Alberto Garcia
2017-04-07 12:41   ` Kevin Wolf
2017-04-07 14:24     ` Alberto Garcia
2017-04-21 21:09   ` [Qemu-devel] proposed qcow2 extension: cluster reservations [was: " Eric Blake
2017-04-22 17:56     ` Max Reitz
2017-04-24 11:45       ` Kevin Wolf
2017-04-24 12:46       ` Alberto Garcia
2017-04-07 12:20 ` [Qemu-devel] " Stefan Hajnoczi
2017-04-07 12:24   ` Alberto Garcia [this message]
2017-04-07 13:01   ` Kevin Wolf
2017-04-10 15:32     ` Stefan Hajnoczi
2017-04-07 17:10 ` Max Reitz
2017-04-10  8:42   ` Kevin Wolf
2017-04-10 15:03     ` Max Reitz
2017-04-11 12:56   ` Alberto Garcia
2017-04-11 14:04     ` Max Reitz
2017-04-11 14:31       ` Alberto Garcia
2017-04-11 14:45         ` [Qemu-devel] [Qemu-block] " Eric Blake
2017-04-12 12:41           ` Alberto Garcia
2017-04-12 14:10             ` Max Reitz
2017-04-13  8:05               ` Alberto Garcia
2017-04-13  9:02                 ` Kevin Wolf
2017-04-13  9:05                   ` Alberto Garcia
2017-04-11 14:49         ` [Qemu-devel] " Kevin Wolf
2017-04-11 14:58           ` Eric Blake
2017-04-11 14:59           ` Max Reitz
2017-04-11 15:08             ` Eric Blake
2017-04-11 15:18               ` Max Reitz
2017-04-11 15:29                 ` Kevin Wolf
2017-04-11 15:29                   ` Max Reitz
2017-04-11 15:30                 ` Eric Blake
2017-04-11 15:34                   ` Max Reitz
2017-04-12 12:47           ` Alberto Garcia
2017-04-12 16:54 ` Denis V. Lunev
2017-04-13 11:58   ` Alberto Garcia
2017-04-13 12:44     ` Denis V. Lunev
2017-04-13 13:05       ` Kevin Wolf
2017-04-13 13:09         ` Denis V. Lunev
2017-04-13 13:36           ` Alberto Garcia
2017-04-13 14:06             ` Denis V. Lunev
2017-04-13 13:21       ` Alberto Garcia
2017-04-13 13:30         ` Denis V. Lunev
2017-04-13 13:59           ` Kevin Wolf
2017-04-13 15:04           ` Alberto Garcia
2017-04-13 15:17             ` Denis V. Lunev
2017-04-18 11:52               ` Alberto Garcia
2017-04-18 17:27                 ` Denis V. Lunev
2017-04-13 13:51         ` Kevin Wolf
2017-04-13 14:15           ` Alberto Garcia
2017-04-13 14:27             ` Kevin Wolf
2017-04-13 16:42               ` [Qemu-devel] [Qemu-block] " Roman Kagan
2017-04-13 14:42           ` [Qemu-devel] " Denis V. Lunev
2017-04-12 17:55 ` Denis V. Lunev
2017-04-12 18:20   ` Eric Blake
2017-04-12 19:02     ` Denis V. Lunev
2017-04-13  9:44       ` Kevin Wolf
2017-04-13 10:19         ` Denis V. Lunev
2017-04-14  1:06           ` [Qemu-devel] [Qemu-block] " John Snow
2017-04-14  4:17             ` Denis V. Lunev
2017-04-18 11:22               ` Kevin Wolf
2017-04-18 17:30                 ` Denis V. Lunev
2017-04-14  7:40             ` Roman Kagan

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=w514ly0s8vw.fsf@maestria.local.igalia.com \
    --to=berto@igalia.com \
    --cc=kwolf@redhat.com \
    --cc=mreitz@redhat.com \
    --cc=qemu-block@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.